The rapid expansion of artificial intelligence is reshaping the global technology industry, with rising demand for AI data centres creating new opportunities for semiconductor companies and other businesses supplying the infrastructure behind the technology.
Marvell Technology has become one of the latest companies to raise its long-term expectations, pointing to growing demand for custom chips and other components used in data centres supporting artificial intelligence.
The company has raised its fiscal 2028 revenue outlook to about $20 billion, reflecting the scale of investment taking place around AI infrastructure.
The development highlights a major shift in the technology market.
Artificial intelligence applications require enormous computing power. Companies developing advanced AI systems need increasingly sophisticated data centres capable of processing huge amounts of information, training models and responding to millions of user requests.
That demand has created a new technology race extending beyond the companies that develop AI software.
Chip designers, semiconductor manufacturers, networking companies, data-centre operators and energy providers are all becoming part of the wider AI infrastructure economy.
Custom chips are particularly important because large technology companies are looking for ways to improve performance while controlling the cost and energy required to operate AI systems.
Instead of relying entirely on general-purpose processors, technology companies can develop specialised hardware designed for particular workloads. These systems can help data centres handle AI applications more efficiently.
The growth of this market has increased the importance of companies such as Marvell, which supplies technologies used in data-centre infrastructure.
The broader semiconductor industry has already benefited from the AI boom.
Companies across the chip supply chain have invested heavily in research, manufacturing capacity and advanced packaging as customers compete for the hardware needed to expand their AI operations.
The scale of spending has also changed the economics of data centres.
AI facilities require large amounts of electricity, advanced cooling systems and high-speed networking equipment. Building them can require billions of dollars in investment, particularly when operators are planning facilities capable of supporting the most demanding AI workloads.
Technology companies are therefore increasingly treating AI infrastructure as a long-term strategic investment rather than a short-term trend.
Major cloud providers have announced enormous spending plans for data centres and computing infrastructure. The objective is to secure enough capacity to support AI services while also preparing for future increases in demand.
For chip companies, that spending can translate into large and long-term orders.
“The AI boom is creating demand for a new generation of computing infrastructure, with specialised chips and powerful…”
Marvell’s updated outlook reflects expectations that demand for AI-related infrastructure will remain strong as more businesses adopt artificial intelligence.
The potential market is also expanding beyond the largest technology companies.
Banks, retailers, manufacturers, healthcare organisations and media companies are increasingly experimenting with AI tools. As adoption grows, demand for computing resources could continue increasing.
However, the expansion of AI infrastructure also creates challenges.
Data centres consume significant amounts of electricity, making access to reliable power a growing concern. In some markets, technology companies are exploring new energy arrangements and locations where electricity supplies can support large computing facilities.Cooling is another challenge.
High-performance AI systems generate substantial heat and require advanced cooling technologies. As computing power increases, data-centre operators must find more efficient ways to manage energy use and heat.
The rapid growth of the sector has therefore created opportunities for businesses far beyond traditional chipmakers.
Companies involved in power generation, electrical equipment, cooling systems, networking and construction are increasingly connected to the expansion of AI infrastructure.
Investors are watching the sector closely because the enormous spending required to build AI capacity has the potential to create a new cycle of growth across the technology industry.
At the same time, companies must prove that the huge investments being made today can eventually generate sustainable returns.
The AI boom has already changed the semiconductor market, but the next stage could be even larger as businesses move from experimenting with artificial intelligence to deploying it at scale.
That transition will require more computing power, more data centres and more specialised hardware.For chipmakers, the opportunity is clear.
If AI adoption continues expanding as expected, demand for the infrastructure supporting the technology could remain one of the most important growth drivers in the global semiconductor industry for years to come.
